A&A 574, A123 (2015) Astronomy DOI: 10.1051/0004-6361/201423830 & c ESO 2015 Astrophysics

Pre-hibernation performances of the OSIRIS cameras onboard the Rosetta spacecraft

S. Magrin1,2, F. La Forgia2, V. Da Deppo3, M. Lazzarin2, I. Bertini1, F. Ferri1, M. Pajola1, M. Barbieri1, G. Naletto1,4, C. Barbieri1,2, C. Tubiana5, M. Küppers6, S. Fornasier7,8, L. Jorda9, and H. Sierks5

1 Centro di Ateneo di Studi e Attivitá Spaziali “Giuseppe Colombo”, University of Padova, via Venezia 15, 35131 Padova, Italy e-mail: [email protected] 2 Department of Physics and Astronomy, University of Padova, vicolo dell’Osservatorio 3, 35122 Padova, Italy 3 CNR-IFN UOS Padova LUXOR, via Trasea 7, 35131 Padova, Italy 4 Department of Information Engineering, University of Padova, via Gradenigo 6, 35131 Padova, Italy 5 Max-Planck-Institut für Sonnensystemforschung, Justus-von-Liebig-Weg 3, 37077 Göttingen, Germany 6 European Space Astronomy Centre, ESA, Villanueva de la Cañada, 28691 Madrid, Spain 7 LESIA, Observatoire de Paris, CNRS, UPMC Univ Paris 06, Univ. Paris Diderot, 5 Place J. Janssen, 92195 Meudon Pricipal Cedex, France 8 Université Paris Diderot, Sorbonne Paris Cité, 4 rue Elsa Morante, 75205 Paris, France 9 Laboratoire d’Astrophysique de Marseille, CNRS and Université de Provence, 38 rue Frédéric Joliot-Curie, 13388 Marseille, France Received 18 March 2014 / Accepted 6 December 2014

ABSTRACT

Context. The ESA cometary mission Rosetta was launched in 2004. In the past years and until the spacecraft hibernation in June 2011, the two cameras of the OSIRIS imaging system (Narrow Angle and Wide Angle Camera, NAC and WAC) observed many different sources. On 20 January 2014 the spacecraft successfully exited hibernation to start observing the primary scientific target of the mission, comet 67P/Churyumov-Gerasimenko. Aims. A study of the past performances of the cameras is now mandatory to be able to determine whether the system has been stable through the time and to derive, if necessary, additional analysis methods for the future precise calibration of the cometary data. Methods. The instrumental responses and filter passbands were used to estimate the efficiency of the system. A comparison with acquired images of specific calibration stars was made, and a refined photometric calibration was computed, both for the absolute flux and for the reflectivity of small bodies of the solar system. Results. We found a stability of the instrumental performances within ±1.5% from 2007 to 2010, with no evidence of an aging effect on the optics or detectors. The efficiency of the instrumentation is found to be as expected in the visible range, but lower than expected in the UV and IR range. A photometric calibration implementation was discussed for the two cameras. Conclusions. The calibration derived from pre-hibernation phases of the mission will be checked as soon as possible after the awak- ening of OSIRIS and will be continuously monitored until the end of the mission in December 2015. A list of additional calibration sources has been determined that are to be observed during the forthcoming phases of the mission to ensure a better coverage across the wavelength range of the cameras and to study the possible dust contamination of the optics. Key words. instrumentation: detectors – space vehicles: instruments – techniques: photometric

1. Introduction In Sect.2 we therefore study the past performances of the two cameras, the Narrow Angle and Wide Angle Camera (NAC The ESA Rosetta mission was launched on 2 March 2004, and and WAC), through the observations of calibration stars. In par- since then, the Optical, Spectroscopic, and Infrared Remote ticular, we investigate the past responses of calibrator signal Imaging System (OSIRIS, Keller et al. 2007), which is a two- through time with different filters to be able to evaluate possible camera system, performed many different scientific observa- aging effects on the optics and to determine whether the pho- tions. It acquired data during a swing-by with Mars and Phobos tometric requirements set by the original mission science goals in 2007, a fly-by with asteroid (2867) Steins in 2008, two swing- have remained stable. We also inspect the dynamical range of the bys with Earth in 2007 and 2009, and a fly-by with asteroid two CCDs by checking the linear dependence between the stellar (21) Lutetia in 2010, to mention only the main activities. On signal and the exposure time of observations. The transmission 8 June 2011, the spacecraft entered into deep-space hiberna- and reflectivity curves of each component of the two cameras, tion mode. It successfully exited hibernation on 20 January 2014 measured during on-ground calibration tests, is used to estimate for the cruise to the rendez-vous with the primary scientific tar- the efficiency of the instrumentation. We thus check and com- get of the mission: the Jupiter-family comet 67P/Churyumov- pare the expected calibrator signals that are deduced from these Gerasimenko. instrumental efficiency curves and from the observed one, which Establishing the stability of the instrument through time is is computed from images acquired in flight. vital to performing a correct absolute flux calibration of the Following this investigation, we present in Sect. 3 the robust cometary data. formulation for the absolute flux calibration of OSIRIS images. Article published by EDP Sciences A123, page 1 of 13 A&A 574, A123 (2015)

Table 1. Basic parameters of the NAC and WAC system and CCD specification (Keller et al. 2007).

NAC WAC Optical design 3-mirror off-axis 2-mirror off-axis Detector type 2k × 2k CCD 2k × 2k CCD Angular resolution (µrad px−1) 18.6 101 Focal length (mm) 717.4 140 (sag)/131 (tan) Field of view (◦) 2.20 × 2.22 11.35 × 12.11 Telescope aperture 6.49 × 10−3 m2 4.91 × 10−4 m2 F-number 8 5.6 Spatial scale from 1km (cm px−1) 1.86 10.1 NAC and WAC CCD Pixel size 13.5 × 13.5 µm2 Full well >120 000 e− px−1 System gain ≈3.1 e−/DU Readout noise (CCD) ≈15 e− rms Dark charge generation <0.1 e− s−1 px−1 @180 K ≈400 e− s−1 px−1 @293 K – (with dithering) Readout rate 1.3 Mpx s−1; 650 kpx s−1 per channel Readout time (full frame) 3.4 s (2 channels)

Table 2. NAC and WAC filter names and wheel positions. plate – visible (NFP-Vis) with respect to far focus plate – visible (FFP-Vis) is also planned, which will allow optimizing the res- NAC WAC olution of the NAC for observations of the comet surface at a Position Wheel 1 Wheel 2 Wheel 1 Wheel 2 distance shorter than 4 km. 1 FFP_UV FFP_IR empty empty We can therefore define the filters as twins, which can be ob- 2 FFP_Vis Orange Green Red tained by combining different clear filters in addition to the same 3 NFP_Vis Green UV 245 UV 375 band-pass filter. These combinations share the same passband, 4 Near-IR Blue CS CN but have different efficiencies in general because clear filters 5 Ortho Far-UV UV 295 NH2 have different efficiencies. Together with defining twin combina- 6 Fe2O3 Near-UV OH Na tions, it is also possible to define a baseline filter combination by 7 IR Hydra UV 325 OI selecting a representative of the group and deduce the flux cali- 8 Neutral Red NH Vis 610 bration of the whole group according to the baseline calibration. Notes. Position of the two filter wheels of all filters for NAC and WAC. In AppendixA we list and define the physical parameters In principle, every combination is possible, but only some of them are and units we used. useful for scientific purposes. 2. Performance of OSIRIS cameras

Section4 is dedicated to the description of the conversion be- 2.1. Expected count rate tween the signal observed on the surface of solar system bodies −1 The expected total count rate, Kx,y in DN s , of a given source and of their correct geometric reflectance. on an image acquired through the (x, y) filter combination is The NAC is a three-mirror anastigmat system, while the expressed by WAC has an off-axis optical configuration obtained with two as- Z ∞ n pherical mirrors (Keller et al. 2007). In Table1 we report the Kx,y = ϕ(λ) · M (λ) · Fx,y(λ) · TARP(λ) main characteristics of the NAC and WAC system, together with 0 the specifications of the CCDs. λ 1 × Q(λ) · · · AP dλ, (1) A plane-parallel antireflection coated plate, referred to as hc IG antiradiation plate (ARP), was added in front of the CCD for where radiation shielding. Both cameras are equipped with two filter wheels placed in – ϕ(λ) is the spectral irradiance of the source at the camera entrance aperture, expressed in W m−2 nm−1; front of the CCD. A passband is defined by a combination of two − filters, hereafter (x, y), where x is the filter position on the first – Q(λ) is the CCD quantum efficiency expressed in e /ph; wheel and y is the position on the second wheel. In Table2 we – IG is the inverse gain, which hereafter is taken to be 3.1 e−/DN; list all filters available for the two cameras by specifying their n position on the wheels. – M (λ), Fx,y(λ), TARP(λ) are the mirror spectral reflectance The WAC filters are located in seven of the eight positions (through n mirrors: n = 2 for WAC or n = 3 for NAC), filters, of the two filter wheels, from 2 to 8, leaving the first position in and antiradiation plate (ARP) transmissions as a function of the wavelength; both wheels empty. All seven filters of the first wheel can be used −4 2 by setting the empty position in the second one, and vice-versa. – AP is the telescope aperture: 4.91 × 10 m for the WAC . × −3 2 In the NAC the filter wheels have no empty position. Neutral and 6 49 10 m for the NAC. or clear filters are used in combination with band-pass filters to In Fig.1 we report the reflectivity of the three-mirror (NAC) optimize the camera performances either in the UV, in the vis- and two-mirror (WAC) systems together with filters and ARP ible, or in the IR range. A different focus range for near focus transmission curves.

A123, page 2 of 13 S. Magrin et al.: Pre-hibernation performances of the OSIRIS cameras onboard the Rosetta spacecraft

Fig. 1. Efficiency curves of NAC and WAC optics: transmission curves of clear and band-pass filters, antiradiation plates (ARP), and reflectivity of mirrors.

The integrand of Eq. (1) is the wavelength dependance of The raw data were processed to remove the bias level. Flat- the count rate detected by the instrument. In particular, Q(λ) × field correction was applied by means of images that were ob- λ/hc is the way to express the CCD efficiency in e−/(W s), where served on the ground of the integrating sphere, which were ac- λ/(hc) = λ/(1.986×10−16) if λ is expressed in nm. Q(λ)×λ/(hc· quired when the cameras were characterized in the laboratory. IG) hence gives the CCD efficiency in DN/(W s). In-flight flat-field images are regularly acquired by illuminating If the source flux is constant with time, then ϕ(λ) is time in- the closed telescope front doors with internal lamps. However, dependent. Since in principle all instrument optics responses de- internal lamps do not guarantee a completely flat source, and pend on age, the integrand of Eq. (1) and hence Kx,y are therefore this method is only useful to monitor the flat-field stability, but time (age) dependent. not to create new reference images. Since no variation was de- tected on in-flight images, laboratory images are still used for the flat-field correction. 2.2. Observed count rate In the WAC images the 16 Cyg A+B pair is not separated, which forced us to apply the method to both stars at the same Equation (1) gives the total count rate that is expected on the time. In the NAC images the 16 Cyg A and B centroids are sepa- CCD in each (x, y) passband of a source with a known spectral rated by about ten pixels. This means that we can measure fluxes irradiance ϕ(λ). As a first check of the OSIRIS performances, we of 16 Cyg B alone only when the aperture radius is verified to be compared these theoretical values with the real values observed smaller than 5 pixels, to avoid contamination from 16 Cyg A. by the instrument. On the CCD, different filters show different typical point Two stars were regularly observed by OSIRIS during the past spread function (PSF) profiles (see for example Figs.2 and3), so mission phases for calibration purposes: Vega and the double that the best aperture radius to use for the aperture photometry star 16 Cyg A+B. 16 Cyg B is a solar analog and is suitable for depends on the filter, which in many cases is larger than 5 pix- measuring the reflectance of small bodies of the solar system. els. For consistency between different cameras and among dif- The corresponding spectral irradiances that were taken into ac- ferent filters, we therefore decided to derive the photometry from count to provide the theoretical values we compared our values both stars in the NAC in the same way as we have to do with with were taken from Bohlin & Gilliland(2004) for Vega and WAC images. See AppendixB for more details on the double from Burlov-Vasiljev et al.(1995, 1998) for the solar spectrum star 16 Cyg A+B. properly scaled to match 16 Cyg A and B. We finally had to define a general criterion to find the best value for the aperture radius. By investigating the stellar integral 2.2.1. Aperture photometry profile obtained at different aperture radius, we typically expect to find a plateau for aperture radii larger than a critical value af- We derived the observed count rate from the calibration stars ter the background level is removed. For OSIRIS images, a typ- with the aperture photometry method. ical aperture growth curve of Vega is reported in Fig.4, where

A123, page 3 of 13 A&A 574, A123 (2015)

In Table3 we report the mean value of the count rate com- puted from Vega images with different filter combinations, the maximum deviation, and the typical exposure times. For the two NAC (F16 FFP-UV_Near-UV, and F82 Neutral_Orange) and two WAC filters (F12 Red, and F18 Vis610) the signal linearity with increasing exposure time was also investigated. The shortest exposure time is imposed by the shutter response tolerance, the longest time is defined by the CCD saturation level. The OSIRIS CCDs are equipped with an antiblooming sys- tem, designed to “extract” overabundant electrons from the pixel potential well. This system is needed for the cometary target to allow overexposed images of the nucleus without saturation ar- tifacts. This is necessary to acquire high signal-to-noise (S/N) Fig. 2. PSF FWHM as number of pixels (geometric average of x and data of the dust and gas in the inner coma against the bril- liant nucleus. The saturation level was required to be about y direction values) in OSIRIS filters deduced from Vega observations. − The largest size of four NAC filters is acquired by combining a near 120 000 e /px ≈ 40 000 DN/px (see Keller et al. 2007; Thomas focus plate that can optimize the performances of the NAC camera for et al. 1998) so that the pixel linearity is guaranteed only be- a target that is at a distance smaller than 4 km. low this limit. Above the saturation, the CCD cannot collect any other electron, and we will find a constant signal regardless of the exposure time. For point sources, which are dominated by the optics PSF, the sky level is estimated in a region here defined by sky aper- the peak value of the source must not exceed the saturation value, tures between 50 and 60 px. For both NAC and WAC, the sky while the total flux of the source is split into more pixels and background is normally found to be within ±3 DN, with a stan- the total number of electrons collected can obviously exceed the dard deviation of about 8.5 DN on NAC and 6.7 DN on the WAC, limit. When the peak signal reaches the saturation levels, the an- probably due to coherent noise. The plateau indicates that the to- tiblooming system prevents overabundant charges of only the tal flux of the star was taken into account. overexposed pixels to spill over into the neighborhood above a In some OSIRIS images this cannot be achieved. We often given exposure time. Nearby unsaturated pixels, however, can found an oblique (instead of a horizontal) asymptote, regardless continue to collect electrons, so that we would still experience of our definition of the sky annuli. We ascribed this to an on- an increase of the total point source signal with exposure time, axis contamination, maybe due to diffuse light by optics micro- even with a lower rate. roughness or to ghosts. To study the linearity of the CCD response, we defined as The NAC ghosts are well described in Dohlen et al.(2010). outliers and consequently excluded from the data set the count First-generation ghosts are caused by the light reflection between rates that deviate from the mean value by more than three times pairs of optical surfaces. The most intense on-axis ghost, for ex- the standard deviation. The iteration of this method defined a ample, can be produced by the reflection of the incident light on time interval where the linearity is satisfied. Figure5 shows the the surface of the CCD, reflected again by the nearest surface distribution of Vega observations (reported as both counts and of the ARP. In the [800−1000] nm IR range the NAC CCD re- count rate) vs. exposure time in this interval. In Table3 the time flectivity increases from 10 to 30%. The ghost relative intensity, interval is specified for these four filters. with respect to the scientific image, is thus three times higher at Table3 shows that almost all the filters are stable within 1000 nm than at 800 nm. The two filters most affected by the ±1.5%. In only one case, WAC-F51 at 295 nm, there is a de- oblique asymptote are the reddest ones, F61 and F71, which are viation of +2.79%, while in only two other filters, NAC-F41 centered at about 930 and 990 nm. at 880 nm and WAC-F61 at 308.5 nm, deviations exeed 1.5%. However, the CCD reflectivity and the ARP reflectivity in the Moreover, we also verified that the data considered do not show whole [250−400] nm range are higher than at 1000 nm, which any evident aging effect since there is no systematic trend with ± means that the ghost problem should affect the UV range much date: the data dispersion of about 1.5% is found both in case of more than the IR, while we find that the contamination in UV stellar observations acquired during the same day and during the NAC filter, where present, is not so evident. Therefore, further whole OSIRIS activity period. This dispersion is also consistent investigation is needed on this topic. with the photometry precision requirement of about 2% set by the original mission science goals (Keller et al. 2007). In the meanwhile, we had to find a new definition for the From this investigation we found that the peak level of counts aperture radius for the NAC-F61 and NAC-F71 filters: we se- on a single pixel in raw images is above 57 000 DN for both lected the smallest aperture radius r such that the flux in the an- NAC filters investigated, while it is about 47 000 and 50 000 DN nulus [r−1, r] is below half the sky standard deviation. After that for the two WAC filters investigated, well above the 40 000 DN we considered the remnant contribution as background even if limit foreseen. We verified for WAC that the centroids of the PSF with a non-zero average. were located in different positions on the CCD, with different efficiencies: if the saturation limit was found to be at 47 000 DN, 2.2.2. CCD linearity and stability the star always fell on some few cold pixels. This difference was then corrected by flat fielding. The difference between NAC and We investigated the stability of the results of the aperture pho- WAC values suggest that the two CCDs behave differently. For tometry method during different observational phases of the mis- NAC-F82 we did not exclude any observation in the data-set sion (from the Mars swing-by in 2007 to the Lutetia fly-by in with the longest exposure times, so that we did not confirme the 2010). saturation limit in this case. For new tests that are scheduled to

A123, page 4 of 13 S. Magrin et al.: Pre-hibernation performances of the OSIRIS cameras onboard the Rosetta spacecraft

a) b)

c) d)

Fig. 3. Shape of the Moffat fit of the PSF of OSIRIS camera filters in the following com- binations: a) NAC-F16 (FFP-UV_Near-UV); b) NAC-F82 (Neutral_Orange); c) WAC-F12 (Red); and d) WAC-F18 (Vis610).

different visual absolute magnitudes, to allow a direct compari- son with the solar spectrum. See AppendixB for further details. The error bars in this plot were computed by propagating the photon noise deduced on the original images with errors defined on each instrument element. Horizontal bars refer to the FWHM of each passband. The total relative (percentage) instrumental error is domi- nated by the CCD QE uncertainty. Several considerations arise from the analysis of the plot:

– In the UV range, that is, below 400 nm, fewer counts are observed than expected. – In the visible range, the WAC filters counts agree with the Fig. 4. Typical curve of the aperture flux growth. This curve was specif- predicted counts. The NAC filter counts, however, seem to F16 ically obtained from Vega observations with the NAC filter. exceed the expected counts (in some cases they are even higher than the vertical error bars). – In the near-IR range, above 800 nm, fewer counts are ob- study the CCD linearity, a wider range in texp for NAC-F82 must served counts than predicted. be considered. – The NAC twin filter values agree within error bars. However, In Fig.3 we show the PSF for two NAC and two WAC filters we found some systematic differences among NAC FFP-Vis, in the case where the centroid of the PSF lies exactly on the NAC NFP-Vis, and NAC Neutral (second, third, and eighth center of a pixel. For the four selected filters the ratios between position on filter wheel 1, respectively). the peak value and the total integrated signal are about 17.5% – The agreement between observed and expected counts is ver- for NAC-F16, 24.5% for NAC-F82, 34.5% for WAC-F12, and ified from both Vega and 16 Cyg A+B counts in the range 34.3% for WAC-F18 of the total signal. 470–890 nm. – The agreement between Vega and 16 Cyg A+B data is good 2.3. Count rate: observed vs. expected above 470 nm. – The scatter among filters and between different stars is larger In Fig.6 we show the ratio between observed and expected in the UV range. 16 Cyg A+B data are systematically lower counts in each filter for Vega and 16 Cyg A+B. The count rate than Vega data. obtained for Vega and 16 Cyg A+B observations acquired dur- – Around WAC F71 (centered on 325 nm) the plot behaves ing all mission phases was compared with the expected rate strangely in both Vega and 16 Cyg A+B data. This scatter that we computed from the Vega reference spectrum (Bohlin is not due to an anomalous observation because the photo- & Gilliland 2004, CALSPEC version 004) and solar reference metric values were computed for this filter, like for others, spectra (Burlov-Vasiljev et al. 1995, 1998), respectively, by as the average of many observation in different phases, giv- means of Eq. (1). For this last source we computed a conversion ing a positive and a negative maximum percentage error of factor of between 16 Cyg A+B and solar fluxes, based on the +1.11% and −1.36%, respectively.

A123, page 5 of 13 A&A 574, A123 (2015)

Table 3. OSIRIS NAC and WAC filter characteristics and Vega observations.

Filter Central λ FWHM Texp Mean count rate +err -err Observations Aperture nm nm s DN/s % % px NAC-F15 267.5 53 0.400 2.571e+005 1.07 0.95 6 6−8 NAC-F16 368.5 33 0.020−0.256 1.415e+006 0.59 0.75 30 8−10* NAC-F22 648.5 83 0.020 7.172e+006 1.19 0.93 5 9−11 NAC-F23 535.5 61 0.020 7.265e+006 0.80 0.69 5 10−12 NAC-F24 481.0 72 0.020 9.369e+006 0.15 0.30 5 10−11 NAC-F27 700.5 21 0.080 1.366e+006 1.29 0.91 5 9−10 NAC-F28 742.0 62 0.025 3.447e+006 0.27 0.22 5 9−11 NAC-F32 648.5 83 0.040 7.496e+006 0.41 0.83 5 13−14 NAC-F33 535.5 61 0.040 7.581e+006 1.38 1.25 5 13−15 NAC-F38 742.0 62 0.060 3.565e+006 1.10 0.61 4 13−14 NAC-F41 880.0 62 0.100 9.906e+005 1.76 1.03 5 10−11 NAC-F51 804.5 39 0.100 9.895e+005 1.16 0.79 5 9−10 NAC-F58 790.5 23 1.000 7.631e+004 0.50 0.34 6 9−10 NAC-F61 929.5 39 0.350 3.223e+005 0.71 1.15 5 13−15 NAC-F71 987.0 38 0.800 1.092e+005 1.43 1.50 5 16−17 NAC-F82 650.5 81 0.070−1.297 1.941e+005 0.58 0.76 32 9−11* NAC-F83 535.5 59 0.500 2.033e+005 1.32 0.80 5 8−9 NAC-F88 742.0 60 0.500 1.752e+005 1.37 1.43 5 9−10 WAC-F12 627.0 154 0.010−0.150 1.169e+006 1.39 1.35 66 8−11* WAC-F13 375.0 8 6.000 1.875e+004 1.23 0.85 8 11−12 WAC-F14 387.0 4 6.000 1.588e+004 1.21 0.93 9 10−12 WAC-F15 571.0 10 1.500 8.142e+004 0.73 0.53 9 9−11 WAC-F16 589.5 3 3.500 3.030e+004 0.77 0.55 8 9−10 WAC-F17 630.5 3 3.500 2.029e+004 0.85 0.97 9 7−9 WAC-F18 611.5 9 0.233−1.900 8.288e+004 0.52 1.14 29 8−10* WAC-F21 535.5 61 0.200 5.910e+005 0.68 0.28 9 9−11 WAC-F31 245.5 13 15.000 5.308e+003 0.85 1.27 9 12−13 WAC-F41 258.0 4 50.000 1.749e+003 1.05 0.84 8 12−14 WAC-F51 295.0 10 25.000 3.060e+003 2.79 1.79 6 12−13 WAC-F61 308.5 3 60.000 1.273e+003 1.77 1.85 8 11−13 WAC-F71 325.5 9 20.000 4.376e+003 1.11 1.36 9 12−13 WAC-F81 335.0 4 80.000 1.422e+003 0.86 0.69 8 11−12

Notes. This table refers to the observations of Vega during different mission phases, from the Mars swing-by in 2007 to the Lutetia fly-by in 2010. For each filter we report the central wavelength of the whole instrumental and filter passband in nm; the FWHM of each instrumental and filter passband in nm; the exposure time needed for the Vega observation, and for the four investigated filters, the linearity interval in seconds; the average count rate in DN s−1 for Vega observations; the positive and negative maximum deviation (expressed in %) with respect to the mean value; the total number of observations; the typical values of the apertures among all observations; (∗) for each of the four additional filters, this refers only to observations performed with a single particular exposure time within the interval specified in the fourth column.

16 Cyg A+B data can be affected by an incorrect computation of method were found to be 7−9 pixels. We recall here that for the conversion from solar fluxes (computed from 10−0.4(VSA−V )). these filters we always found a plateau. However, throughout all However, the match of 16 Cyg A+B data with Vega data above the past mission phases, the signal was always found to increase 470 nm suggests that this assumption is good at least in the vis- for the NAC IR filters F61 and F71; it lay on an oblique asymp- ible and near-infrared range of the spectrum. In the UV range, tote. Taking into account a possible on-axis contamination, we there may be also some differences due to the imperfect match decided to stop at an aperture of about 10−12 pixels for F61 and between 16 Cyg A and 16 Cyg B spectra and the solar spec- at 15−17 pixels for F71, depending on the background noise trum, which can be variable in these wavelengths by up to 20% level on the different images. For the sake of completeness, we (Colina et al. 1996). Moreover, below 325 nm, 16 Cyg A+B investigated wider apertures for the NAC-F61 and F71 filters. counts are maybe unreliable as a result of the decreasing sig- We found that even 40-pixel apertures are not enough to obtain nal in this wavelength range that is typical of solar analog stars, the expected counts, hence wider apertures cannot improve the which means that we can only trust the Vega counts. calibration in either case. Variability among twin filters that share the same passband (i.e., NAC F82-F22-F32, NAC F83-F23-F33, NAC F88-F28- F38, etc.) are probably due to different shapes of the actual trans- 3. Absolute flux calibration mission curves of clear filters (NAC FFP-Vis, NAC NFP-Vis, We call the factor needed to convert observed count rates per and NAC Neutral) with respect to laboratory data of the same pixel into physical units the abscal factor (i.e., spectral radiance, filters obtained on the ground before the launch. expressed in W m−2 nm−1 sr−1) to obtain the absolute instrument The lower count rate in the near IR can in principle be due calibration. to an incorrect computation of the counts within the selected We present photometric calibration factors for each pass- aperture. For many NAC filters the aperture radii selected for band. We used images of stellar targets such as Vega and the the computation of Vega flux through the aperture photometry pair 16 Cyg A+B. According to each filter combination (x, y),

A123, page 6 of 13 S. Magrin et al.: Pre-hibernation performances of the OSIRIS cameras onboard the Rosetta spacecraft

Fig. 5. Linearity measurements obtained by the Vega observations in terms of DN or DN/s vs exposure time. For each filter, the upper panels show counts vs. exposure time in the particular filter, the lower panels show the count rate vs. the exposure time. The investigation was made with the following filters: a) NAC-F16 (FFP-UV_Near-UV); b) NAC-F82 (Neutral_Orange); c) WAC-F12 (Red); and d) WAC-F18 (Vis610).

Fig. 6. Ratio between observed and expected counts from the CCD with a particular filter setting. Vertical error bars refer to photon noise and the propagation of instrumental uncertainties. Horizontal bars refer to the FWHM of each passband. Red data refer to counts computed from 16 Cyg A+B images and solar spectrum (Burlov-Vasiljev et al. 1995, 1998) scaled to 16 Cyg A+B magnitude; black data refer to Vega images and Vega spectrum (Bohlin & Gilliland 2004).

we can measure the total count rate due to the star (or pair), Kx,y, the CCD expressed in steradians (3.547e-010 sr for NAC and from these images. 9.982e-009 sr for WAC): The spectral radiance that we expect on each pixel with the R +∞ x, y ϕ λ Mn λ F λ T λ Q λ λ AP λ same filter combination ( ) is the weighted average of the in- 0 ( ) ( ) x,y( ) ARP( ) ( ) hc IG d coming spectral radiance of the target star, with the instrumen- φx,y = R +∞ · pxsz Mn(λ) F (λ) T (λ) Q(λ) λ AP dλ tal response as the weight, divided by the pixel size (pxsz) of 0 x,y ARP hc IG

A123, page 7 of 13 A&A 574, A123 (2015)

Table 4. Absolute flux calibration of the OSIRIS NAC and WAC filters.

Filter Central λ FWHM Abscal factor +err -err Abscal reflectance factor +err -err nm nm DN s−1 · (Wm−2 nm−1 sr−1)−1 % % DN s−1 %% NAC F15 267.5 53.0 2.428e+006 16.7 16.7 1.434e+005 15.3 15.3 NAC F16 368.5 33.0 1.308e+007 5.6 5.7 4.566e+006 5.6 5.7 NAC F21 659.0 492.0 6.190e+008 3.3 3.2 2.830e+008 1.7 1.7 NAC F22 648.5 83.0 1.182e+008 2.5 2.4 5.892e+007 2.5 2.4 NAC F23 535.5 61.0 6.677e+007 2.4 2.5 4.01e+007 2.4 2.5 NAC F24 481.0 72.0 6.421e+007 2.7 2.5 4.086e+007 2.7 2.5 NAC F26 368.5 33.0 1.232e+007 5.6 5.7 4.307e+006 5.6 5.7 NAC F27 700.5 21.0 2.824e+007 2.4 2.4 1.270e+007 2.4 2.4 NAC F28 742.0 62.0 8.544e+007 2.3 2.3 3.495e+007 2.3 2.3 NAC F31 653.5 479.0 6.297e+008 3.2 3.2 2.907e+008 3.9 3.9 NAC F32 648.5 83.0 1.235e+008 2.5 2.4 6.156e+007 2.5 2.4 NAC F33 535.5 61.0 6.959e+007 2.5 2.6 4.169e+007 2.5 2.6 NAC F34 481.0 72.0 6.741e+007 2.6 2.5 4.290e+007 2.6 2.5 NAC F35 289.5 13.0 3.157e+005 14.9 14.9 3.980e+004 14.5 14.5 NAC F36 371.5 27.0 1.071e+007 5.6 5.7 3.765e+006 5.5 5.6 NAC F37 700.5 21.0 2.942e+007 2.4 2.4 1.323e+007 2.4 2.4 NAC F38 742.0 62.0 8.842e+007 2.3 2.3 3.618e+007 2.3 2.3 NAC F41 880.0 62.0 4.024e+007 3.4 3.2 1.209e+007 3.4 3.2 NAC F51 804.5 39.0 3.159e+007 2.7 2.5 1.141e+007 2.7 2.5 NAC F58 790.5 23.0 2.326e+006 2.9 2.5 8.626e+005 2.9 2.5 NAC F61 929.5 39.0 1.444e+007 4.3 4.3 3.894e+006 4.3 4.3 NAC F71 987.0 38.0 5.763e+006 10.8 10.7 1.384e+006 10.7 10.6 NAC F81 722.0 206.0 1.937e+007 3.1 3.1 8.938e+006 3.8 3.8 NAC F82 650.5 81.0 3.246e+006 2.7 2.4 1.607e+006 2.7 2.4 NAC F83 535.5 59.0 1.869e+006 2.5 2.4 1.120e+006 2.5 2.4 NAC F84 481.0 72.0 1.991e+006 2.7 2.5 1.270e+006 2.7 2.5 NAC F86 378.0 18.0 4.980e+004 5.1 5.2 1.819e+004 5.0 5.1 NAC F87 700.5 21.0 1.313e+006 2.4 2.4 5.891e+005 2.4 2.4 NAC F88 742.0 60.0 4.286e+006 2.7 2.8 1.756e+006 2.7 2.8 WAC F12 627.0 154.0 4.810e+008 2.5 2.6 2.52e+008 2.5 2.6 WAC F13 375.0 8.0 4.684e+006 4.8 4.8 1.774e+006 4.8 4.8 WAC F14 387.0 4.0 2.538e+006 3.8 3.9 8.826e+005 3.8 3.8 WAC F15 571.0 10.0 2.557e+007 2.3 2.1 1.494e+007 2.3 2.1 WAC F16 589.5 3.0 1.045e+007 2.1 2.1 5.823e+006 2.1 2.1 WAC F17 630.5 3.0 8.594e+006 2.1 2.3 4.435e+006 2.1 2.3 WAC F18 611.5 9.0 3.196e+007 2.1 2.2 1.724e+007 2.1 2.2 WAC F21 535.5 61.0 1.523e+008 2.3 2.1 9.131e+007 2.3 2.1 WAC F31 245.5 13.0 1.370e+006 23.6 23.5 3.449e+004 20.9 20.7 WAC F41 258.0 4.0 4.686e+005 13.9 13.8 1.952e+004 12.9 12.8 WAC F51 295.0 10.0 8.483e+005 11.8 11.7 1.443e+005 11.7 11.6 WAC F61 308.5 3.0 3.592e+005 9.8 9.7 7.412e+004 9.6 9.5 WAC F71 325.5 9.0 1.312e+006 8.6 8.6 3.690e+005 8.6 8.5 WAC F81 335.0 4.0 4.447e+005 7.7 7.6 1.357e+005 7.7 7.6

Notes. In this table we report the absolute flux and reflectance calibration of all OSIRIS NAC and WAC filter combinations, obtained using the whole data set of Vega and 16 Cyg A+B observations during the different mission phases from the Mars swing-by in 2007 to the Lutetia fly-by in 2010. For each filter we report the central wavelength of the whole instrumental and filter passband in nm; the FWHM of each instrumental and filter passband in nm; the abscal factor needed to convert the image count rate into spectral radiance; its positive and negative percentage error; the abscal reflectance factor needed to convert the image count rate into reflectance; its positive and negative percentage error.

−1 −2 −1 −1 −1 Consequently, the abscal factor, Ax,y, which links the total count The abscal factor is expressed in DN s · (Wm nm sr ) . rate (Kx,y) on the CCD for a particular filter combination to In Table4 we report the abscal factor values for all di fferent the spectral radiance within the passbands, φx,y, expressed in filters computed from this equation. W m−2 nm−1 sr−1, is defined by Q(λ) is roughly constant on the visible passbands and can be simplified from both integrals in Eq. (3): this means in par- Ax,y = Kx,y/φx,y. (2) ticular that the abscal factors are little affected by the imperfect In first approximation Ax,y is filter dependent (and hence in knowledge of the CCD quantum efficiency in the visible range principle time (age) dependent) and star independent. (approximately [400−1000] nm). Because AP, hc, and IG are constant with wavelength and can be simplified, the abscal factor is R +∞ 3.1. Twin combinations K Mn λ F λ T λ Q λ λ λ x,y pxsz 0 ( ) x,y( ) ARP( ) ( ) d Ax,y = R +∞ · (3) As said in the Introduction, we can calibrate scientific im- ϕ(λ) Mn(λ) F (λ) T (λ) Q(λ) λdλ 0 x,y ARP ages acquired with any of the filters of a twin group by using

A123, page 8 of 13 S. Magrin et al.: Pre-hibernation performances of the OSIRIS cameras onboard the Rosetta spacecraft

Table 5. List of NAC filters without stellar calibration. In this form we can see that the abscal factor should be calibrator independent. However, the expected counts in general may be Camera Filter Baseline not equal to the observed counts, as shown in Sect. 2.3. This is to calibrate filter why the previous formulation is only the theoretical formulation NAC 26 16 and needs to be corrected for each variation on the instrumental NAC 31 21 response with time (age). NAC 34 24 In the following, we call Cx,y the ratio between observed and NAC 35 15 expected counts in the (x, y) filter. It is defined by NAC 36 16 NAC 37 27 Kx,y observed NAC 81 21 Cx,y = R +∞ · (5) ϕ λ Mn λ F λ T λ Q λ λ AP λ NAC 84 24 0 ( ) ( ) x,y( ) ARP( ) ( ) hc IG d NAC 86 16 NAC 87 27 This equation defines the correction factor that we need to apply to the optics profiles so that Eq. (1) gives the observed count rate. Notes. To calibrate these filters, we used the count rate from the baseline However, with this definition we only assume single correc- filters listed in the right column. tion factors, constant on wavelengths, for each (x, y) filter combination. In this approximation the abscal factor we should use to the baseline calibration of a representative filter of the group. convert our images becomes For example, we can calibrate both NAC-F87 and NAC-F37, or NAC-F84 and NAC-F34, simply by using the baseline calibra- Ax,y real  Cx,y Ax,y theoretic. (6) tion obtained with filter combinations NAC F27 or NAC F24, respectively. In Table5 we show all the filters for which this is We stress that this is not an exact equivalence. We should not possible and the selected baseline filter. only correct for Cx,y factors, constant on wavelength, but we In the following we describe the equations used to properly should find a correction profile C(λ), dependent on the wave- infer the absolute calibration in these cases. As an example, we length, deduced by Cx,y data, that can correct the shape of every use the calibration of NAC-F84 images through the NAC-F24 passband profile. observation of the calibration star. The filter in the second po- However, where the response diversity is a constant factor sition (2) on the NAC first filter wheel (FFP-Vis: far focus over the passband (or maybe even a slow symmetrical variation plate – visible) has a mean efficiency ≈95%, while filter 8 over the passband), the theoretical source flux weighted mean is (neutral) has ≈5%. Accordingly, the two considered filters re- the same as the observed one, and the previous equivalence is fer to Neutral_Blue or FFP-Vis_Blue (coded as NAC-F84 or exact. This is verified for OSIRIS filters in the [470−900] nm and NAC-F24), respectively. To compute the abscal factor of the range. filter combination A84, we cannot use the usual expression In this case, we can write

R +∞ n K M λ F λ T λ Q λ λ λ Ax,y real = Kx,y observed pxsz 84 pxsz 0 ( ) 84( ) ARP( ) ( ) d A84 = R +∞ R +∞ n ϕ λ Mn λ F λ T λ Q λ λ λ M (λ) Fx,y(λ) TARP(λ) Q(λ) λdλ 0 ( ) ( ) 84( ) ARP( ) ( ) d 0 × ∞ · (7) R + n ϕ(λ) M (λ) Fx,y(λ) TARP(λ) Q(λ) λdλ since we did not observe calibration stars through the F84 filter, 0 we do not know the K count rate, but only K . However, we 84 24 Therefore, we can convert the count rate into the correct physical expect from Eq. (1) that the counts of a star through a filter are units in each passband (x, y) because the abscal factor compen- proportional to the integral of the stellar spectrum multiplied by sates for every variation in the optics efficiencies: the passbands, so that Kx,y expected Kx,y observed R +∞ n K ϕ(λ) M (λ) TARP(λ)Q(λ)F84(λ)λdλ φx,y =  84 0 Ax,y theoretic Ax,y real = R +∞ · (4) K24 ϕ(λ) Mn(λ) T (λ)Q(λ)F (λ)λdλ 0 ARP 24 deduced from Eq. (2).

This means that we can express A84 as In contrast, where differences between actual and laboratory measured profiles are asymmetric over the passbands, Eq. (7) is R +∞ Mn λ F λ T λ Q λ λ λ valid only in a first approximation, and the goodness of the ap- pxsz 0 ( ) 84( ) ARP( ) ( ) d A84 = K24 R +∞ · proximation depends on the real shape of the passbands. In this ϕ λ Mn λ F λ T λ Q λ λ λ 0 ( ) ( ) 24( ) ARP( ) ( ) d case we could also find different absolute flux calibrations for every source observed. This may be the reason why we see a This method was applied to complete the absolute flux calibra- mismatch between Vega and 16 Cyg A+B data in the UV range tion presented in Table4 for filters without a direct observation in Fig. 6. The steep UV behavior of the plot suggests that in this of the calibration stars. range, NAC and WAC passbands may have profiles with a shape different from that considered for the computation. These differ- 3.2. Theoretical and real abscal factor ent shapes give an incorrect weight for computing the weighted arithmetic means that is at the base of the concept of Eq. (3). Since the expected counts from a source should be expressed as These considerations are of fundamental importance for in- in Eq. (1), the theoretic abscal factor should also be expressed as vestigating possible aging or dust contamination of the optics Z +∞ during the last part of the mission around the comet, since we ex- n λ AP Ax,y theoretic = pxsz M (λ) Fx,y(λ) TARP(λ) Q(λ) dλ· pect a wavelength-dependent variation of optical efficiencies in 0 hc IG these cases. Therefore, we will have to continuously check this

A123, page 9 of 13 A&A 574, A123 (2015) calibration: the computation of abscal factors may change with where Ix,y is the body intensity at zero phase angle, ϕ x,y(r) is the time, and the calibration will be good only because these fac- incident solar spectral irradiance at the body distance r from the tors compensate for efficiency variations. Moreover, an accurate Sun, and the factor π comes from the integral on the semi-sphere computation of the correction profile C(λ) deduced from Cx,y of the Lambert cosine law. must be implemented to improve the UV calibration. We can accordingly define the reflectance R of a solid object One way to monitor the goodness of the approximation in illuminated by the Sun at a given phase angle α as Eq. (6) is to compute, for example, the flux calibration by means of different sources, with different spectra ϕ(λ). Ix,y(α) Rx,y(α) = , (9) ϕ x,y(r)/π 3.3. List of UV calibrators where Ix,y(α) is the intensity emitted by the body in the (x, y) We should therefore observe different stars with different spec- band at the phase angle α and ϕ x,y(r) is the peak of the solar tral type to be able to check the consistency of the calibration flux diffused by the Lambertian disk with the same cross-section independently of the star spectrum so that we can try to fix cali- of the object at the distance r from the Sun in the same band. bration issues after the awakening of OSIRIS. As we stated be- From the solar spectrum at 1 AU (Burlov-Vasiljev et al. fore, 16 Cyg A and B are solar analogs, with a very low S/N in 1995, 1998) we can compute the expected solar count rate at the UV, which is the reason why at the moment only Vega is used 1 AU, according to Eq. (1): at these wavelengths. We need stars different from Vega (A0V) Z +∞ n λ AP and hotter than a solar analog (G2V). Kx,y = ϕ (λ) M (λ) Fx,y(λ) TARP(λ) Q(λ) dλ, In Table6 we present a list of stars in the Bruzual- 0 hc IG Persson-Gunn-Stryker (BPGS) atlas with spectral type earlier −1 expressed in DN s , where ϕ (λ) is the solar spectral irradi- than A3. These stars are cross-correlated with the TD1 catalog −2 −1 (Thompson et al. 1978) and the corresponding magnitudes are ance at 1 AU expressed in W m nm . The solar spectral ir- also provided (Hayes & Latham 1975). radiance at 1 AU inside a particular filter (x, y) is expressed by The BPGS atlas consists of 175 spectra of stars covering a the weighted mean: complete range of spectral types and luminosity classes. Each in- R +∞ ϕ (λ) Mn(λ) F (λ) T (λ) Q(λ) λdλ dividual spectrum covers the [229–25 600] Å wavelength region. 0 x,y ARP ϕx,y = R +∞ · The fluxes were arbitrarily normalized to a visual magnitude of Mn λ F λ T λ Q λ λ λ 0 ( ) x,y( ) ARP( ) ( ) d zero. More details about the spectra can be found in Gunn & Stryker(1983). After dividing it by π and scaling it to the distance r of the target to the Sun expressed in astronomic units, we obtain the −2 −1 −1 value ϕ (r)/π, expressed in W m nm sr , used to obtain 4. Reflectance calibration the (geometric) reflectance within a given passband. Note that the reflectance is a adimensional number. If we are observing a small body in the solar system, at the first approximation and at least in the visible range, we detect that the As already cited, we used the spectrum of the Sun measured solar spectrum is diffused from the body surface. Considering by Burlov-Vasiljev et al.(1995, 1998). Solar analogs can be sig- OSIRIS data, we can remove the solar contribution on each nificantly different from the Sun, especially in the UV range, image by computing the solar flux that falls inside each filter where the Sun can be variable by up to 20% (Colina et al. 1996). passband to obtain the spectrophotometric reflectance of the ob- See AppendixB for more details. served object inside each passband. We are able, therefore, to In the following we use the expression investigate its optical properties and give hints on its surface R +∞ ϕ (λ) Mn(λ) F (λ) T (λ) Q(λ) λ dλ mineralogical composition. ϕx,y 0 x,y ARP = R +∞ (@1 AU), For example, we wish to scale the image count rate so that if π π · Mn λ F λ T λ Q λ λ λ 0 ( ) x,y( ) ARP( ) ( ) d the target is observed at zero phase angle, it will give the value (10) of the geometric albedo on the given passband. For that purpose, we recall some photometric definitions. or The geometric albedo is the ratio of a body brightness ϕ K at zero phase angle to the brightness of a perfectly diffusing x,y x,y = R +∞ , (Lambertian) disk with the same position and apparent size as π π · Mn λ F λ T λ Q λ λ AP λ 0 ( ) x,y( ) ARP( ) ( ) hc IG d the body. A Lambertian surface follows the cosine emission law (Lambert 1760): the intensity seen by an observer from an ideal to indicate the mean spectral irradiance of the Sun at a distance diffusing surface is proportional to the cosine of the angle θ be- of 1 AU in the wavelength range defined by the passband (x, y). tween the observer’s line of sight and the normal to the surface, Since the star observed by OSIRIS cannot be directly the Sun, with peak luminous intensity (Ipeak) in the normal direction. The but is a solar analog, the counts on the CCD refer to this second incident solar flux is equal to the total diffused flux that can be star. We can thus infer the “solar” counts from the analog counts computed from the peak intensity by integrating the cosine law (Kx,y SA) by using on the semi-sphere defined by the normal direction, obtaining −0.4(V −VSA) Ftot = π Ipeak. Kx,y = Kx,y SA10 , The geometric albedo in a (x, y) given passband is expressed by so that the reflectance factor is

−0.4(V −V ) ϕx,y Kx,y 10 SA Ix,y SA p = , (8) = R +∞ , (11) x,y π π · Mn λ · F λ T λ Q λ λ AP λ ϕ x,y(r)/π 0 ( ) x,y( ) ARP( ) ( ) hc IG d

A123, page 10 of 13 S. Magrin et al.: Pre-hibernation performances of the OSIRIS cameras onboard the Rosetta spacecraft Cep β m1565 m1965 m2365 m2740 V mag B mag U 0.30 5.310 6.209 6.243 3.103 4.184 4.067 3.558 0.22 5.640 6.094 5.727 6.232 5.726 6.443 − − 01.205004.3314 00.2996 0.4918.202400.1028 5.110 0.6018.4248 2.09 5.93326.3204 5.160 2.1431.2650 4.240 3.56 5.931 5.98115.2527 4.230 5.56 5.15006.4397 3.380 7.17 5.966 5.21715.2064 3.380 2.42 5.340 4.19100.1416 2.940 6.41 3.048 5.417 4.10001.4940 5.110 5.64 1.187 4.355 3.63057.0274 5.90 4.081 1.427 4.300 5.82025.4782 1.58 2.827 10.33 0.632 3.80007.1328 4.490 3.996 2.924 12.88 0.595 5.98010.5126 2.255 2.147 0.249 6.830 4.90008.3118 1.54 3.538 2.327 2.120 2.479 5.79808.3674 1.96 1.800 1.653 1.739 6.810 4.99112.6461 4.860 3.33 1.967 1.555 3.917 5.886 7.22109.8229 4.870 2.77 4.381 1.221 1.224 5.263 2.487 5.62236.4457 3.64 4.716 1.132 3.477 3.351 7.282 5.72322.1582 5.720 3.47 4.381 0.790 6.106 3.629 10.12 5.68006.3974 4.660 4.754 3.041 4.481 5.216 5.893 6.37409.9588 5.810 6.04 3.611 5.880 3.287 3.020 5.450 5.14301.0307 7.16 3.204 4.132 6.143 2.189 6.498 6.19601.4153 0.36 4.007 5.478 2.888 3.804 5.362 5.249 5.310 1.88 4.058 3.720 5.898 3.552 3.452 6.287 4.890 3.883 5.294 3.679 3.261 2.476 5.485 6.090 3.786 5.463 3.046 4.699 3.724 6.154 5.860 3.514 5.273 3.576 4.301 3.638 5.533 6.255 3.420 2.721 4.326 4.896 6.132 5.940 5.235 4.458 3.310 3.726 6.152 3.907 4.488 4.742 2.623 5.011 4.056 2.912 4.677 5.045 4.089 5.501 3.608 4.902 4.402 5.306 5.336 3.270 4.531 4.033 5.680 4.937 2.947 5.397 − − − − − + + + − + + − + + − + + − + + + + + − − + + (deg) (deg) mas filters in Johnson system. m1565, m1965, m2365, m2740: magnitudes from TD1 survey at 1565, 1965, V , B , U 24 21 38.626918 40 18.750944 55 29.0048 6.0089 56.4824 46 09 85.6967 20.781620 48 52.407903 23 87.1548 55.1867 47.4189 223.2510 15 50 17.520638 15 60.1618 58.281019 46 24.246232 47 68.5130 22.8336 55.3515 12 45 69.4690 32.684440 35 59.137847 55 31.1700 54.461107 56 75.2156 22.261038 23 81.8068 01.7740 28.2285 35 11 71.9693 09.446833 34 05.761319 15 67.5696 18.062416 47 56.1333 20.979236 40 43.9685 02.583234 25 55.5157 22.464447 25 83.3947 03.5627 71.6918 87.5108 46 00 22.8001 72.3219 05 32 20.4185 311.4176 36 06 00.5654 67.2324 − + + + + + + + + + − + + − + + + + + + + + + − + : magnitudes in V , B , ... 13 09 56.99067 ... 19 50 37.32728 U + + number; TD1: TD1 number; name: common name; type: object classification in Simbad; SpT: spectral type; RA, Dec: right ascension and declination;  Hya bC* B3V 08 43 13.47499 Cap *i* B9IV 20 20 39.81562 Vir *i* A1IVs Her bC* B3IV 17 39 27.88609 Lyr Be* B6IV 19 07 18.12878 η ι θ ν ι HIP88 469 TD197 796103 37125 Name 25 980 593 27 499103 732 9 Sge88 4826 886 HR 8023 27 586 942 Sgr 799 Type86 21 414 BD-01 752 60 935 Cyg * SpT101 12 909 946 EB* 102 Her84 V* 20 605 820 27 077 Em*93 104 O7.5Ia... Be* O6V((f))96 20 275 178 B1IV HR O4V((f)) 7899 *i*98 23 441 639 19 B1Ve 52 20 21.76387 38 56 Oph64 34.77848 24 238 977 RA(ICRS) 18 03 HR7174 52.44466 B2IV *i*100 25 310 869 05 32 9 41.35338 Vul97 16 634 666 21 01 26 HD 510 10.92841 189689 *i*97 B3V 774 Al* Dec(ICRS) -01 35 18 30.590594 25 08 385 579 45.49142 Be*96 B9.5V 25 491 640 B7IV 205.1376 V* HR 756793 24 903 126 B9 20 Glon 39 HR 04.96819 759195 25 067 114 20 B8IIIn 17 Aql83 17 23 313 989 39.53024 Al* 18 58 HR 01.89799 746787 24 192 509 V* Glat98 -26 19 234 37 866 B1III 20 44.4788 19 00 34 07.05730 HR 34.89569 7346105 21 811 V* 062 * B2III 59 358.5545 Her99 25 145 767 Plx 28 HR 096 6642103 312 B3V *i* 11 mag Sge 26 B5II-III 103 69 Cyg 27 19 484 52 07.16659 * * B9V HR 7699 HR 19 8020 36 19 56.64950 12 40.71241 * Pu* A3IV A1V * 19 Pu* 20 33.05554 B0Ib B9III B8Iae 17 B5Ib 01 36.36149 17 48 47.87388 21 25 19 47.02562 57 45.44583 20 55 49.80412 20 07 41.44099 Selected UV calibrators from the BPGS catalog. Column description: HIP: H Table 6. Notes. 2365, 2740 angstroms. Description oftype; types. Pu*: * pulsating : variable star; star; *i*: Be*: star Be in star; double Em*: system; emission-line EB*: star. eclipsing binary; Al*: eclipsing binary of Algol type (detached); V*: variable Star; bC*: variable Star of GLON, GLAT: Galactic coordinates; Plx: parallax; mag

A123, page 11 of 13 A&A 574, A123 (2015) where the use of the V magnitude, at each wavelength range, Sun spectrum corrects the difference caused by the two different for the computation of the factor 10−0.4(V −VSA) is justified since spectral types. we assume the same spectral shape for the Sun and the solar The two methods are expected to give the same values. If analog. Some authors refer to the value of the solar flux as al- the absolute calibration is reliable and if the abscal factors are ready divided by π. To avoid confusion, we always explicitly the same regardless of the calibration star used, we can choose write ϕx,y /π as the solar intensity at 1 AU. one of the two formulations to compute the best reflectance fac- We can compute the reflectance directly from raw images tor just according to the smallest error bars associated with the (bias and flat-field corrected) by defining a composed abscal re- result. flectance factor, multiplying the abscal factor defined in Eq. (3) Vega is a single star and has a much better S /N than 16 Cyg and the new reflectance factor as expressed in Eq. (10). A or B in each filter. After checking that the calibration is con- Hence, the abscal reflectance factor defined is the value sistent when computed using different calibrators, we found that needed to convert the count rate into the reflectance of a body the best star to use is Vega, both for the abscal factor computation located at 1 AU from the Sun: and for the solar values computation. This formulation was verified to be valid. Good matches, ϕx,y Kx,y calstar · pxsz Ax,y · = within error bars, between OSIRIS spectrophotometry and lit- π π erature reflectance spectra can be found for Phobos (Pajola et al. R +∞ ϕ λ Mn λ F λ T λ Q λ λ λ 2012, 2013) and Lutetia (Magrin et al. 2012). 0 ( ) ( ) x,y( ) ARP( ) ( ) d × ∞ R + n ϕ(λ)calstar M (λ) Fx,y(λ) TARP(λ) Q(λ) λdλ 0 4.2. Twin combinations (@1 AU), (12) When we have to calibrate NAC F84 images from NAC F24 where the “calstar” source (calibration star, usually Vega) used baseline acquisition, we have to substitute Eq. (4) into Eq. (12), for the absolute flux calibration is generally different from the finding “SA” source (solar analog, usually 16 Cyg A+B, or 16 Cyg B) used to compute the solar value in Eq. (11). In Table4 we re- ϕ84 K24 calstar · pxsz A · = port the abscal reflectance factor values computed by means of 84 π π Eq. (12). R +∞ n ϕ (λ) · M (λ) · F (λ)T (λ)Q(λ)λdλ To derive the value needed for a body at a heliocentric 0 84 ARP × R +∞ 2 ϕ λ · Mn λ · F λ T λ Q λ λ λ distance r, we simply have to divide this factor by r if r is 0 calstar( ) ( ) 24( ) ARP( ) ( ) d expressed in AU. @1 AU.

4.1. Counts from the solar analog This method was applied to complete the absolute reflectance calibration presented in Table4 for filters without direct In the visible range the S/N of 16 Cyg B for the WAC is better observation of the calibration stars. than 100 with 180 s exposure time even with narrowband filters. For the NAC, the S/N is better than 200 with an exposure of a few seconds from about 480 nm to 1000 nm. In this spectral 5. Conclusions range we can also use this solar analog as a flux calibration star. We detected a good stability within ±1.5% of the instrumental Equation (12) can thus be further simplified by using “SA” as performances, without evidence of aging effects on the optics “calstar”, giving from 2007 to 2010. We found a good match between observed

−0.4(V −VSA) and expected data as deduced from on-ground laboratory mea- ϕx,y Kx,y SA × pxsz × 10 Ax,y · = @1 AU, (13) sured optics and CCD efficiencies in the visible range. In the UV π π and IR wavelength range, however, we noted lower instrumental

−0.4(V −VSA) efficiencies than expected. since we used ϕ = ϕSA10 . This leads to smaller error bars because we simplified every A reliable formulation for the flux calibration was proposed instrumental passband; the passbands are the parameters carry- to be able to monitor this behavior from just after the exit from ing the largest uncertainties. Moreover, through this method we hibernation of the cameras and during the whole journey toward can exclude some problems that are due to the instrumental vari- and around the comet, to study the possible dust contamination ation with time (age) and with some uncertainties in the optics of the optics that would imply a wavelength-dependent variation efficiencies because they all affect the target and stellar images of optical efficiencies. We also presented the formulation for the in the same way. reflectance calibration. This formulation was verified to be valid, However, we have to remember that there may be some dif- since there are good matches between OSIRIS spectrophotom- ferences between the Sun and the solar analog, especially in the etry and literature reflectance spectra for Phobos (Pajola et al. UV, where the calibration with both 16 Cyg A+B may lead to 2012, 2013) and Lutetia (Magrin et al. 2012). further approximations, and that this solar analog is quite faint While the calibration compensates for instrumental effi- in this wavelength range so that we have to work with a low S/N ciency variations in the visible and IR range, a deeper inves- in the UV filters. tigation of UV filters must be implemented. For this purpose, UV counts of 16 Cyg A+B have large error bars because we finally suggested a list of other calibration sources. This will the spectrum is very weak in these wavelengths. In this case, ensure a good calibration across the whole wavelength range the only way to find a reflectance value is to return to Eq. (12), during the entire last part of the mission. computing the count rate from a star like Vega (which is much Acknowledgements. OSIRIS was built by a consortium led by the Max-Planck- brighter than the Sun, especially in the UV). The division by the Institut für Sonnensystemforschung, Katlenburg-Lindau, Germany, in collabora- weighted Vega spectrum and the multiplication by the weighted tion with CISAS, University of Padova, Italy, the Laboratoire d’Astrophysique

A123, page 12 of 13 S. Magrin et al.: Pre-hibernation performances of the OSIRIS cameras onboard the Rosetta spacecraft

Table B.1. Properties of the Sun, 16 Cyg A, and 16 Cyg B according to Takeda(2005) and Takeda et al.(2005).

Star Sp. type Teff V π MV B.C. log(L/L ) L/L mass Sun G2V 5778 −26.74 4.83 1 HD 186408 16 Cyg A G1.5V 5765.1 5.99 46.2 4.32 −0.13 +0.24 1.74 1.02 HD 186427 16 Cyg B G2.5V 5795.4 6.25 46.7 4.60 −0.12 +0.10 1.26 1.03

Notes. The HD number and the conventional stellar name for the three stars; the spectral type; the effective temperature derived from the spec- troscopic approach reported in Takeda et al.(2005); the apparent V magnitude and the trigonometric parallax π in milliarcsecond, taken from the H Catalog (ESA 1997); the absolute V magnitude, MV , derived from the π and V values; the bolometric correction obtained from the results reported in Alonso et al.(1995); stellar luminosity log( L/L ) and L/L derived in Takeda et al.(2005) from MV and B.C.; stellar mass (in units of M ) determined by L/L and Teff . de Marseille, France, the Instituto de Astrofisica de Andalucia, CSIC, Granada, If placed at the same distance as the Sun, the flux of 16 Cyg Spain, the Research and Scientific Support Department of the European B is about 1.26 times the solar flux, and the flux of the pair Space Agency, Noordwijk, The Netherlands, the Instituto Nacional de Tecnica 16 Cyg A+B is about three times the solar flux. Aeroespacial, Madrid, Spain, the Universidad Politechnica de Madrid, Spain, the Department of Astronomy and Space Physics of Uppsala University, Sweden, In summary: 16 Cyg A and B are slightly different from the and the Institut für Datentechnik und Kommunikationsnetze der Technischen Sun, but they seem to be similar enough to the Sun to allow us Universität Braunschweig, Germany. The support of the national funding agen- to assume the same spectral shape for the three stars, at least to cies of Germany (DLR), France (CNES), Italy (ASI), Spain (MEC), Sweden the first order. Then, the simplest way to compute the scaling (SNSB), and the ESA Technical Directorate is gratefully acknowledged. This ac- tivity has been realized under the WAC/Rosetta Agenzia Spaziale Italiana (ASI) factor to convert the flux of 16 Cyg A+B into the solar flux is to contract to the Istituto Nazionale di Astrofisica (INAF) I/024/12/0. We also thank consider V16 CygA = 5.99, V16 CygB = 6.25, and V = −26.74 at the anonymous referee of the manuscript for providing constructive comments 1 AU. The magnitude V of the pair is then that helped to improve this paper. −0.4·V16 CygA −0.4·V16 CygB, V16 Cyg A+B = −2.5 · log10 10 + 10. Appendix A: Units and the difference between the magnitude V of the 16 Cyg pair and the Sun leads to the relation In the following we list and define the physical parameters and −0.4·(V16 CygAB−V ) units we used. ϕ16 Cyg A+B = ϕ · 10 – The irradiance is the power incident on a surface. It is because we assume this factor to be a constant along the spec- −2 expressed in the International System as W m . trum, since we assume the same spectral shape for the three stars. – The spectral irradiance is the power incident on a surface per To estimate the count rate of the Sun from the count rate of −2 −1 unit wavelength, here expressed as W m nm . 16 Cyg A+B, the relation is – The radiance is the power emitted by a source per unit solid −0.4×(V16 Cyg AB−V ) angle and unit area perpendicular to the emitting direction, Kx,y = Kx,y 16 Cyg A+B/10 expressed as W m−2 sr−1. – The spectral radiance is the power per unit solid an- at 1 AU, as we wrote in previous paragraphs. gle per unit area per unit wavelength, here expressed as W m−2 nm−1 sr−1. References The abscal factor presented in this paper is the factor needed to convert images from count rate into spectral radiance, so that the Alonso, A., Arribas, S., & Martinez-Roger, C. 1995, A&A, 297, 197 pixel-per-pixel signal associated with the target, observed in a Bohlin, R. C., & Gilliland, R. L. 2004, AJ, 127, 3508 Burlov-Vasiljev, K. A., Gurtovenko, E. A., & Matvejev, Y. B. 1995, Sol. Phys., particular passband, is directly comparable with a flux calibrated 157, 51 spectrum of the target in a wavelength interval around the central Burlov-Vasiljev, K. A., Matvejev, Y. B., & Vasiljeva, I. E. 1998, Sol. Phys., 177, wavelength of the band. 25 Colina, L., Bohlin, R. C., & Castelli, F. 1996, AJ, 112, 307 Dohlen, K., Jorda, L., Lamy, P., Toth, I., & Origne, A. 2010, SPIE Conf. Ser., Appendix B: Description of the double 7731 Gunn, J. E., & Stryker, L. L. 1983, ApJS, 52, 121 star 16 Cyg A+B Hayes, D. S., & Latham, D. W. 1975, ApJ, 197, 593 Keller, H. U., Barbieri, C., Lamy, P., et al. 2007, Space Sci. Rev., 128, 433 The two stars 16 Cyg A and B are slightly different from the Lambert, J. H. 1760, Photometria, sive, De mensura et gradibus luminis, colorum Sun. In particular, 16 Cyg A is larger than 16 Cyg B, and both et umbrae (Augustae Vindelicorum: V.E. Klett) are larger than the Sun. From Takeda(2005) and Takeda et al. Magrin, S., La Forgia, F., Pajola, M., et al. 2012, Planet. Space Sci., 66, 43 Pajola, M., Lazzarin, M., Bertini, I., et al. 2012, MNRAS, 427, 3230 (2005) we can find the log L of both 16 Cyg A and B, in terms Pajola, M., Lazzarin, M., Dalle Ore, C. M., et al. 2013, ApJ, 777, 127 of L , transformed from the absolute visual magnitude MV by Takeda, Y. 2005, PASJ, 57, 83 means of the bolometric correction (B.C.) obtained by interpo- Takeda, Y., Ohkubo, M., Sato, B., Kambe, E., & Sadakane, K. 2005, PASJ, 57, lating the results reported in Alonso et al.(1995). We summarize 27 these results in Table B.1, where T is computed for all stars Thomas, N., Keller, H. U., Arijs, E., et al. 1998, Adv. Space Res., 21, 1505 eff Thompson, G. I., Nandy, K., Jamar, C., et al. 1978, Catalogue of stellar ultravio- from a spectroscopic method using the equivalent width values let fluxes A compilation of absolute stellar fluxes measured by the Sky Survey of FeI and FeII lines. Both 16 Cyg A and B are very close to the Telescope (S2/68) aboard the ESRO satellite TD-1 Sun.

A123, page 13 of 13